ZMIME
Comparison · 3 models · Updated Oct 4, 2026

Kimi K2 Thinking vs Llama 4 Scout 17B Instruct vs Mistral Small 3.1 24B

Llama 4 Scout 17B Instruct comes out ahead, 62 to 58 and 55 on our weighted score, though Mistral Small 3.1 24B is 18% cheaper per token.

  1. Moonshot AI

    Kimi K2 Thinking

    Released Nov 6, 2025

    58/100
    • ECI146.0
    • Price$0.60 / $2.50
    • Context262K
    Full specs Remove Kimi K2 Thinking
  2. Our pick

    Meta

    Llama 4 Scout 17B Instruct

    Released Apr 5, 2025

    62/100
    • ECI129.7
    • Price$0.225 / $0.69
    • Context10M
    Full specs Remove Llama 4 Scout 17B Instruct

  3. Mistral AI

    Mistral Small 3.1 24B

    Released Mar 17, 2025

    55/100
    • ECI127.5
    • Price$0.229 / $0.436
    • Context128K
    Full specs Remove Mistral Small 3.1 24B
01 — Verdict

Llama 4 Scout 17B Instruct is our pick

Llama 4 Scout 17B Instruct is the better all-round choice, scoring 62/100 against Kimi K2 Thinking (58) and Mistral Small 3.1 24B (55). It leads on context window. Kimi K2 Thinking wins on capability. Mistral Small 3.1 24B wins on price and inputs & features. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityKimi K2 ThinkingCapabilities Index (ECI): Kimi K2 Thinking 146.0 · Llama 4 Scout 17B Instruct 129.7 · Mistral Small 3.1 24B 127.5
  • Lowest priceMistral Small 3.1 24BMistral Small 3.1 24B $0.281 · Llama 4 Scout 17B Instruct $0.341 · Kimi K2 Thinking $1.07 per 1M tokens (3:1 blend)
  • Longest contextLlama 4 Scout 17B InstructLlama 4 Scout 17B Instruct 10,000,000 · Kimi K2 Thinking 262,144 · Mistral Small 3.1 24B 128,000 tokens
  • Widest inputsLlama 4 Scout 17B Instruct and Mistral Small 3.1 24BKimi K2 Thinking: Text · Llama 4 Scout 17B Instruct: Text, Images · Mistral Small 3.1 24B: Text, Images
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightKimi K2 ThinkingLlama 4 Scout 17B InstructMistral Small 3.1 24B
CapabilityCapabilities Index (ECI)50%735250
Price25%487276
Inputs & features15%355060
Context window10%3710024
Overall100%58/10062/10055/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

Kimi K2 Thinking vs Llama 4 Scout 17B Instruct vs Mistral Small 3.1 24B specifications side by side
SpecificationKimi K2 ThinkingMoonshot AILlama 4 Scout 17B InstructMetaMistral Small 3.1 24BMistral AI
Capability
Capabilities Index (ECI)146.0 (best)129.7127.5
ECI rank#72 of 148 (best)#126 of 148#132 of 148
GPQA DiamondGraduate-level science questions84.2% (best)51.8%47.5%
OTIS Mock AIME 2024–2025Competition mathematics83.1% (best)7.8%5.8%
Price per million tokens
Input$0.60$0.225 (best)$0.229
Output$2.50$0.69$0.436 (best)
Cached input———
Blended (3:1)$1.07$0.341$0.281 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 10 providersMedian of 4 providersMedian of 2 providers
Limits
Context window262,144 tokens10,000,000 tokens (best)128,000 tokens
Max output262,144 tokens (best)16,384 tokens16,384 tokens
Inputs and features
TextYesYesYes
ImagesNoYesYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesNoNo
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsOpenOpenOpen
API model ID———
API providers10 (best)42
ReleasedNov 6, 2025Apr 5, 2025Mar 17, 2025
Knowledge cutoffAug 2024Aug 2024Jun 2024
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • Kimi K2 Thinking$11.00
  • Llama 4 Scout 17B Instruct$3.63
  • Mistral Small 3.1 24B$3.16
04 — Questions

Which should you choose?

Which is better: Kimi K2 Thinking, Llama 4 Scout 17B Instruct or Mistral Small 3.1 24B?

Llama 4 Scout 17B Instruct is the better all-round choice, scoring 62/100 against Kimi K2 Thinking (58) and Mistral Small 3.1 24B (55). It leads on context window. Kimi K2 Thinking wins on capability. Mistral Small 3.1 24B wins on price and inputs & features. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, Kimi K2 Thinking, Llama 4 Scout 17B Instruct or Mistral Small 3.1 24B?

Mistral Small 3.1 24B is cheaper at $0.229 input / $0.436 output per million tokens (median across 2 API providers). Llama 4 Scout 17B Instruct costs $0.225 input / $0.69 output per million tokens (median across 4 API providers); Kimi K2 Thinking costs $0.60 input / $2.50 output per million tokens (median across 10 API providers). At a typical mix of three input tokens to one output token, that is $0.281 per million tokens for Mistral Small 3.1 24B versus $0.341 for Llama 4 Scout 17B Instruct (1.2× as much) and $1.07 for Kimi K2 Thinking (3.8× as much).

Which scores higher on benchmarks?

Kimi K2 Thinking scores higher on the Capabilities Index (ECI): Kimi K2 Thinking 146.0 (#72 of 148), Llama 4 Scout 17B Instruct 129.7 (#126 of 148) and Mistral Small 3.1 24B 127.5 (#132 of 148). Their confidence ranges do not overlap (143.4–147.6 vs 124.8–131.4), so the gap is a real one. On individual benchmarks: GPQA Diamond — Kimi K2 Thinking 84.2%, Llama 4 Scout 17B Instruct 51.8%, Mistral Small 3.1 24B 47.5%; OTIS Mock AIME 2024–2025 — Kimi K2 Thinking 83.1%, Llama 4 Scout 17B Instruct 7.8%, Mistral Small 3.1 24B 5.8%.

Which is better for coding?

There are no published SWE-bench Verified results for Kimi K2 Thinking, Llama 4 Scout 17B Instruct and Mistral Small 3.1 24B yet, so there is no like-for-like coding score. On overall capability, Kimi K2 Thinking leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

Llama 4 Scout 17B Instruct has the largest context window at 10,000,000 tokens, against 262,144 for Kimi K2 Thinking and 128,000 for Mistral Small 3.1 24B. Maximum output per response: Kimi K2 Thinking up to 262,144, Llama 4 Scout 17B Instruct up to 16,384, Mistral Small 3.1 24B up to 16,384 tokens.

Which can read images, PDFs, audio or video?

Kimi K2 Thinking accepts text; Llama 4 Scout 17B Instruct accepts text and images; Mistral Small 3.1 24B accepts text and images. Llama 4 Scout 17B Instruct handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights, so you can self-host them.

Which is newer?

Kimi K2 Thinking is the newest, released Nov 6, 2025. Llama 4 Scout 17B Instruct came out Apr 5, 2025; Mistral Small 3.1 24B came out Mar 17, 2025. Knowledge cutoff: Kimi K2 Thinking Aug 2024, Llama 4 Scout 17B Instruct Aug 2024, Mistral Small 3.1 24B Jun 2024.

How do you decide the winner?

Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.

05 — Keep comparing

Related comparisons